Knowledge Graph

Definition

A knowledge graph is a structured representation of entities and their relationships, stored as nodes and edges in a graph database. It encodes real-world knowledge for reasoning and search.

Purpose

The purpose is to organize knowledge in a machine-readable way. It enables semantic search, recommendations, and reasoning over relationships.

Importance

  • Improves search accuracy through context.
  • Supports explainability in AI systems.
  • Enables integration of structured and unstructured data.
  • Requires ongoing updates to remain accurate.

How It Works

  1. Identify entities (people, places, concepts).
  2. Define relationships between entities.
  3. Populate graph with data from structured/unstructured sources.
  4. Store in a graph database with schema.
  5. Query graph for reasoning or search tasks.

Examples (Real World)

  • Google Knowledge Graph: improves search relevance.
  • Wikidata: open knowledge base for linked data.
  • Microsoft Academic Graph: represents research publications.

References / Further Reading

  • Hogan et al. “Knowledge Graphs.” ACM Computing Surveys.
  • W3C RDF Standard.
  • Google Knowledge Graph documentation.